Forecasting EB-2/EB-3 Green Card Filing Dates - Machine Learning Model

In this blog post, we'll explore the process of forecasting Green Card filing dates using a simple linear regression model in Python. By analyzing historical data from the United States Citizenship and Immigration Services (USCIS), we can use basic machine learning techniques to predict future filing dates. I will walk you through the process step-by-step.

Gathering Data:

   To begin our journey, we need to gather relevant data. You can collect data from USCIS or other trustworthy sources. This dataset should include essential information such as the visa category, country of chargeability, and the final action date for each month. For this use case, I collected data manually from USCIS visa bulletin for India EB-2 and EB-3 categories. Data looks like this -

Visa bulletin -

Building the Linear Regression Model:

   Using Python libraries like scikit-learn, we can construct our linear regression model. This simple yet powerful algorithm will help us forecast Green Card filing dates based on historical data. 

Loading Data -

Preparing Models -

Making Predictions:

   Once our model is trained and evaluated, we can utilize it to make predictions on new data. These predictions will give us estimated filing dates based on the input variables.

EB-2 Prediction -

EB-3 Prediction -

For now, I will stop here and not going to add steps for model validation, accuracy and error metrics. 

You can see, it is so easy to predict the filing date on paper or code. But truth is that nobody knows when we EB-2/EB-3 people get green card.. Thanks to USCIS for speedy process.. Anyways, let me know if you need the data that I used.

And if you want to know when will your filing date will be current, please let me know your priority date in comments.